Ilford FP4 Plus 125

Research-backed original, derived from the Ilford FP4 Plus datasheet (archived, Nov 2018). Ilford's medium-speed fine-art standard since 1968, and, on the numbers, the most forgiving film in this entire bank. Derivation in the notes. Movie-mode translation in the notes.

The number that defines this film

"FP4 Plus is and will give usable results even if it is overexposed by as much as six stops, or underexposed by two stops." Ilford, FP4 Plus technical information, p.1

Eight stops of usable latitude, and it is asymmetric, three times more tolerance for over than under. That is the largest explicitly quantified latitude claim of any datasheet in this collection, larger even than XP2 Super's EI 50-800 single-roll range, and it is the single fact this recipe is built around.

Two things follow directly, and they are the recipe's two most consequential values:

• The protection belongs on the highlight end, because that is where Ilford puts the tolerance. Hence DR400: the maximum highlight headroom the X-T5 offers, with Highlight −1 on top.
• The shadows are ordinary. Two stops of underexposure tolerance is unremarkable. The sheet makes no shadow-specific claim at all. Hence Shadow 0: neutral, not the lifted value the film's reputation would suggest. See the notes for why I followed the sheet rather than the folklore here.

In practice: when in doubt, give it light. Underexposure is the one direction this film does not forgive.

Why the other values

• Acros base. The house standard for every B&W original here, and the datasheet earns it specifically: FP4 Plus is sold for "high print quality", "particularly when giant enlargements are to be made", and for copying, internegative, scientific, technical and industrial work. That is a precision, high-detail, full-tonal-scale film, and Acros is Fujifilm's premium B&W profile. Richer tonality, finer perceived grain structure, better detail than plain Monochrome. (The bank's attributed reference recipe uses Monochrome instead, deliberately and defensibly, see the variants below.)
• Grain Weak, Small. Ilford's opening line is "exceptionally fine grain, medium speed", and PERCEPTOL appears in the developer table as the "finest grain" option. This puts FP4 in the bank's finest tier alongside Acros 100, Delta 100, Pan F Plus 50 and XP2 Super 400. (The reference recipe's Weak/Large is the one grain combination no datasheet supports, fine-grained films have small grain.)
• Sharpness +1. The sheet claims resolution three ways. Giant enlargements, copying/internegative work, and ID-11 at 1+3 listed for "maximum sharpness": so this goes positive. It stops at +1 rather than Pan F Plus 50's +2, because that recipe's higher tier is earned by explicit "edge contrast" language in its datasheet, which this one does not contain.
• Clarity 0, and this is the recipe's clean correction. Same discipline as Delta 100: there is no micro-contrast, edge-contrast or acutance language anywhere in this sheet, so the bank's default holds. The attributed reference recipe runs Clarity +2. That is unsupported by the document. Bonus: HEIF-safe, no shot-to-shot saving penalty.
• ISO 500-3200. DR400 pins the floor at 500 on the X-T5, a deliberate trade, discussed in the notes. The ceiling is modest because Ilford's own quality range tops out at EI 200, and settings above EI 400 appear in the sheet only under the heading "accidental exposure only."
• Exposure Compensation 0 to +2/3. The six-stops-over claim, applied. Not as aggressive as XP2 Super's +1/3 to +1, that film's grain gets finer with exposure, which is a positive reason to overexpose. Here overexposure is merely safe, not beneficial, so the default stays moderate and the latitude is held in reserve for when a scene needs it.
• WB Auto 0/0, Monochromatic Color 0, High ISO NR −4: house standards. On the WB specifically: whether a shift does anything on a monochrome base is currently contested in this bank (five recipes carry large deliberate shifts, one derived from spectral data), so the originals hold at neutral pending a test. See White Balance & WB Shift, the colour-cast control.

The other number worth knowing: this is the bank's best long-exposure B&W

Ilford publishes the reciprocity-failure correction as a formula:

Ta = Tm^1.26: and "for exposures between ½ and 1/10,000 second, no adjustments are needed."

Both halves matter. A ½-second threshold before any correction is needed is generous, and the 1.26 exponent is the lowest of any film in this bank with a published formula, XP2 Super is 1.31. Compounded over a long exposure that gap is not academic:

At the five-minute mark FP4 finishes seven minutes sooner. For night, architecture, water and ND-filter work that is the difference between one usable frame and two, and it is a documented physical property of the emulsion, not a preference.

Nothing about that changes a digital setting. It is recorded because it tells you what this film was for, and because if you shoot a scene on both the film and this recipe, this is the constraint the film was working under.

Variants

Handling

• Give it light, and stop worrying. Six stops of overexposure latitude means there is almost no bright scene this recipe cannot hold at +2/3 with DR400. The failure mode is the other direction.
• It is the ** recipe, not the characterful one. Ilford's own words: "very in processing and will tolerate less than ideal processing conditions." FP4 was engineered to be reliable, and the recipe inherits that. If you want the film to have an opinion, shoot Tri-X or Delta 3200.
• This is the recipe for detail. Copying, architecture, texture, landscape, still life, anything you would print large. It is the closest thing in the B&W set to a technical instrument.
• Where it sits: Pan F Plus 50 is slower, sharper and more etched (Clarity +2, from its own datasheet's edge-contrast language). Delta 100 is the T-grain precision alternative at a similar speed. Acros 100 is the most scan-validated fine-grain recipe here. FP4 is the forgiving one: the fine-grain film you use when the light is difficult.
• Video: see the notes. This recipe translates unusually well. Grain is the only unavailable control doing meaningful work, and at Weak/Small its loss is small.

Test priority

1. A scene with a genuinely extreme brightness range. Bright window plus dark interior, or open sun plus deep shade. This is the film's entire documented identity and the one claim a single frame can settle. Does DR400 + Highlight −1 hold six stops of highlight, or does it clip?
2. A deliberate two-stop underexposure. The asymmetry claim runs the other way too: Ilford says two stops under is still usable. If shadows fall apart sooner than that, Shadow 0 is wrong and should go negative, which would vindicate the reference recipe's reading.
3. A large-detail subject. Brickwork, foliage, fabric weave, to check whether Sharpness +1 with Clarity 0 delivers the "giant enlargements" claim, or whether the reference recipe's Clarity +2 was doing necessary work after all.
4. A face. FP4 has no published skin-tone data, and this recipe's precision settings are not obviously kind to complexions. Worth knowing before recommending it for people.

Validation tier: Datasheet-validated (archived, Ilford FP4 Plus technical information, "exceptionally fine grain", the EI 50-200 quality range, the Ta = Tm^1.26 reciprocity formula with a ½-second threshold, full development tables across 14 developers, the ILFOTEC HC 1+31 characteristic curve). Characteristic pass: consistent with FP4's six-decade reputation as the reliable medium-speed fine-art standard. No real-film scan yet.

What this recipe corrects, and why it matters. The bank has carried an attributed FP4 recipe since before this derivation, and the 2026-07-15 audit already flagged its Grain Weak/Large as "the one grain combination no datasheet supports." Reading the sheet confirms that and adds two more: Clarity +2 is unsupported (no edge-contrast language anywhere), and the Shadow −1.5 models the film's reputation rather than its published latitude, which is asymmetric in the other direction. Those are three corrections from one document that had been sitting unread in this collection since 2026-07-15.

Pro tip: FP4 is the bank's insurance policy (2026-07-25). If you are walking into light you cannot predict or control, this is the B&W recipe to have loaded. Eight stops of documented latitude, the gentlest reciprocity characteristic here, and a manufacturer who says outright that the film "will tolerate less than ideal processing conditions." It is not the most exciting recipe in this folder. It is the one that comes back with an usable frame.

ISO 500 is the sweet spot, not a compromise (measured 2026-07-27). DR400 pins the ISO floor at 500, and that is exactly where the X-T5's conversion-gain step sits, so it is the cleanest ISO in that region, not a tax you pay for using DR400. ISO 640 is the expensive next click: roughly 0.35 EV of dynamic range for one third-stop, against about 0.02 EV for the step from 400 to 500. In good light, hold ISO at 500 and control exposure with aperture and shutter. Measured from published read-noise and PDR data, see Dynamic Range & Tone Curve, the contrast controls.

Movie-mode adaptation of the stills recipe. Target: standard film-simulation recording (baked-in look), not F-Log/HLG. Set in the movie shooting display → IMAGE QUALITY SETTING tab. These are stored separately from stills, so every value must be re-entered.

This is the cleanest photo→video translation in the bank

Movie mode drops Grain, Color Chrome Effect, Color Chrome FX Blue and Clarity (Video / Movie Mode, settings, differences, and photo→video translation video-mode-settings). For most recipes that is a serious loss. Here it is almost nothing:

So the video version is a near-1:1 reproduction of the stills recipe, which is true of almost nothing else here. Every colour recipe loses the Chrome effects. Every gritty B&W recipe loses its grain. The soft creative looks lose Clarity, which is often their entire identity.

Two structural reasons this one survives, worth generalising:

1. Monochrome bases are immune to two of the four losses. Color Chrome Effect and CCFXB act on saturated colour. On Acros they do nothing in stills either, so nothing is lost. (A useful corollary: the X-Trans IV→V blue-conversion rule is irrelevant to all video recipes, since CCFXB is unavailable regardless.)
2. A recipe that already sits at Clarity 0 and fine grain has nothing to give up. This bank's B&W originals that deviate. Tri-X at Clarity +3, Pan F Plus 50 at +2, Delta 3200 at +1, all translate worse.

Practical consequence: if you shoot both stills and video on one job and want them to cut together, this recipe matches itself across the two better than any other in the bank. That is a real production argument for choosing it, independent of how the film looks.

Interframe NR OFF, deliberately

The one value that differs from the stills recipe's spirit.

Interframe NR is temporal noise reduction. It averages across frames, and it can cause ghosting on motion. On a recipe whose entire premise is detail and precision (giant enlargements, copying, technical photography), smearing moving edges is the wrong trade. OFF.

The counter-case: at the ISO 500 floor that DR400 forces, a X-T5 has very little noise to reduce, so Interframe NR is buying almost nothing here anyway. Turn it to AUTO only if you push ISO high in genuinely dim conditions and see noise you cannot live with, and accept the ghosting.

The DR400 / ISO 500 constraint bites harder in video

In stills the ISO 500 floor is a mild inconvenience. In video it is a real constraint, because you also have a fixed shutter angle to respect, the 180° rule puts you at 1/50 s for 25p or 1/60 s for 30p, and you cannot shorten the exposure to compensate for bright light the way you can with stills.

In bright daylight at ISO 500 and 1/50 s you will need ND filters. That is not a fault of the recipe. It is the standard video exposure triangle, and any DR400 recipe has the same issue. Plan for a variable ND or a 6-stop fixed.

If you have no ND available, drop to DR200 (ISO floor 250) and set Highlight −2 to partly compensate for the lost headroom. You lose some of the six-stop latitude that is the point of this recipe, but a correctly-exposed frame beats a faithful one that is blown.

Where video makes this recipe better than its neighbours

Worth stating plainly, because it is not obvious: the loss of Grain in movie mode narrows the gap between the B&W recipes, and FP4 wins on what's left.

In stills, Tri-X, HP5, Delta 400 and FP4 are clearly different films, and much of that difference is grain and Clarity. In video, grain is gone and Clarity is gone, so those four collapse toward each other, differing only in tone curve and Sharpness.

What FP4 keeps that the others do not is its latitude architecture: DR400 plus Highlight −1 plus a documented six stops of overexposure tolerance. In a medium where you cannot bracket, cannot easily reshoot a moving subject, and are locked to a shutter angle, latitude is worth more than character. This is the B&W video recipe for unpredictable light.

For B&W video that wants to look degraded, add grain in post. It is a texture, not a colour, so adding it later does not compromise the SOOC colour goal, and start from Tri-X's tone curve instead.

Notes

• Everything must be re-entered. Movie IQ settings are stored separately from stills. Changing the stills Film Simulation does not change the movie one.
• DR200/400 are only selectable when MOVIE SETTING > F-Log/HLG RECORDING is set appropriately: check this first if DR400 is greyed out.
• If you shoot F-Log / F-Log2 for grading in post, treat this as a target look rather than a bake-in recipe. The latitude argument above applies even more strongly there. F-Log2 captures considerably more range than DR400, but the film-sim character is then applied in the grade, not the camera.
• The variants in the stills recipe transfer directly except where they touch Clarity or Grain. The Acros+Ye / +R filter variants work fully in movie mode, since the filter is part of the film simulation rather than a separate control.

Datasheet-derived original. Primary source: Ilford FP4 Plus technical information, Nov 2018. Archived at the datasheet archive, 6 pages, read in full. Settings in the notes. Video in the notes.

What the datasheet says

Finding 1: this is the widest documented latitude in the bank, and it is asymmetric

"overexposed by as much as six stops, or underexposed by two stops."

Eight stops of usable range, weighted three-to-one toward overexposure. For comparison, against the other explicitly-quantified latitude claims in this collection:

Note the two claims are different in kind and should not be conflated. XP2's is a rating claim. Shoot the whole roll at any EI in that band and process normally. FP4's is a frame-level robustness claim, an individual mis-exposed frame still prints. FP4's is the larger number and the more forgiving property in the field.

The asymmetry is the actionable part, and it is what separates this recipe from the attributed one. Ilford is not saying "FP4 is forgiving." It is saying FP4 forgives you in one specific direction, and the direction is more light.

How that becomes settings

Dynamic Range is a highlight-protection control. It works by underexposing the sensor and lifting the shadows back, buying headroom at the top. That maps exactly onto a six-stop overexposure tolerance, so DR400, the maximum the X-T5 offers. Is the direct translation, with Highlight −1 softening the shoulder on top of it.

And the shadow end gets nothing, because the sheet gives it nothing. Two stops of underexposure tolerance is unremarkable. Most conventional B&W films manage that, and there is no shadow-detail, open-shadow or long-toe language anywhere in the six pages.

The shadow decision

This is the recipe's most contestable value and it deserves the argument in full, because it departs from both the attributed reference recipe (Shadow −1.5) and from FP4's widely-repeated reputation.

The reputation: FP4 "rarely goes fully black" and has "excellent shadow detail" That phrasing appears in the reference recipe's own description and across the photography community. It would argue for a lifted, negative Shadow value.

What the datasheet actually supports: nothing of the kind, explicitly. What it says is that the film tolerates two stops under, which is normal. Against six over. If anything, the document points the opposite way: this is a film whose margin lives at the top of the curve.

So the recipe runs Shadow 0. Neutral. Not lifted, not deepened.

I could be wrong about this, and the test that would prove it is specified in the notes: shoot a deliberate two-stop underexposure. If the shadows hold gracefully, if the film's toe really is long, then the reputation is describing something real that Ilford simply did not write down, Shadow 0 is too conservative, and the reference recipe's −1.5 was right on evidence the sheet doesn't carry. That is a falsifiable claim and it should be tested rather than argued.

Until then, the discipline this bank applies elsewhere holds: the datasheet outranks the folklore, and where the datasheet is silent the default stands. It is the same standard that gave Delta 100 Clarity 0 despite marketing prose that sounded like Pan F's.

The Monochrome-plus-lifted-shadows reading is preserved as a named variant, credited to the reference recipe, so nothing is lost if the test goes the other way.

The DR400 decision, and its cost

DR400 pins ISO to a minimum of 500 on the X-T5. On a film rated ISO 125, that is a two-stop mismatch, and it is worth naming rather than glossing.

Precedent in the bank runs both ways: Delta 100, Acros 100 and Pan F Plus 50 all use DR200 (ISO floor 250) at similar or slower speeds. Following that convention would put FP4 at DR200.

The reason to break it: the convention is a proxy for "slow film, narrow latitude," and FP4 is the counterexample, a slow film with the widest documented latitude in the collection. Setting it to DR200 would model the speed while ignoring the specification, which inverts the bank's own priority. The same reasoning put XP2 Super on DR400.

Why the mismatch matters less than it looks: the ISO number in a digital recipe does not simulate film speed. Grain does, and grain is set independently at Weak/Small. ISO 500 on a X-T5 is essentially noise-free, so the floor costs nothing visible. What it costs is shutter/aperture flexibility in bright light, which is a real but manageable constraint. Anyone shooting wide open in full sun should use the DR200 variant and accept two stops less highlight headroom.

Finding 2: the best long-exposure characteristic in the bank

Ilford publishes reciprocity-failure correction as an explicit formula, which most manufacturers do not:

Ta = Tm^1.26, with "no adjustments needed" from ½ second to 1/10,000 second.

Two comparable data points exist in this collection, both from Ilford sheets read in this collection:

A lower exponent means less compensation required, compounding as exposures lengthen:

This does not change a single digital setting, the X-T5 has no reciprocity failure, and it is recorded honestly as such. It matters for two reasons. First, it tells you what the film was for: architecture, interiors, night, ND-filter water work, the technical and scientific applications the sheet names. Second, if you ever shoot a comparison frame on real FP4, this is the constraint the film is working under and the recipe is not.

The XP2 comparison is the more interesting one, and it inverts an assumption: the chromogenic film with the famously huge rating range has the worse long-exposure behaviour of the two. Latitude and reciprocity are independent properties, and this bank now has the numbers to say so.

Base and Clarity: same standard, applied again

Acros over Monochrome. The reference recipe's Monochrome choice is defensible and I said so in its analysis, a flatter base suits a gentle, unhurried film. Reading the datasheet changes the reasoning: FP4 is sold on "high print quality", "giant enlargements", and copying, internegative, scientific, technical and industrial work. That is a precision instrument, not a gentle one, and Acros. Fujifilm's premium B&W profile, with richer tonality and better detail than plain Monochrome. Is the correct base for it. The Monochrome reading survives as a variant.

Clarity 0. The sheet contains no micro-contrast, edge-contrast or acutance language across six pages. The bank's standard is explicit and has now been applied four times, Pan F Plus 50 earns Clarity +2 because its sheet says "edge contrast". Delta 100, XP2 Super and now FP4 stay at 0 because theirs do not. The reference recipe's +2 is unsupported by the document.

Sharpness +1, not +2. The resolution claims are real. Giant enlargements, copying, and ID-11 (1+3) listed for "maximum sharpness" in the developer table, so it goes positive. But Pan F's +2 tier is tied to that edge-contrast language, and consistency means FP4 lands one notch below.

What this derivation corrects

Three corrections from one document that had been archived in this collection, unread, since 2026-07-15:

The exposure-compensation reversal is the largest disagreement, and it is worth being explicit about. The reference recipe runs negative EC on the reasoning that "slight underexposure protects highlights and builds contrast" and matches "the sheet's tight EI 50-200 latitude."

Reading the sheet, that premise does not hold. The EI 50-200 band is the quality range, not the latitude, the latitude is separately specified as six stops over and two under. Underexposing a film whose documented tolerance is three times better in the other direction is moving toward its weak end, not its strong one. The 2026-07-15 audit accepted the negative-EC convention as. That acceptance was based on the EI range and should be revised now the actual latitude specification has been read.

Where it sits in the B&W set

Ten B&W originals, and the Ilford conventional-grain line is now complete: Pan F 50 → FP4 125 → HP5 400. That sits alongside the tabular Delta line (100/400/3200) and the chromogenic XP2, three emulsion technologies from one manufacturer, all datasheet-derived, which is a genuinely unusual thing for a recipe bank to hold.

Validation status

Datasheet-validated, no scan. Every setting traces to the archived sheet, and the four places this recipe disagrees with the attributed reference are argued against specific document language above rather than asserted.

Highest-value evidence to add, in order:

1. A frame with an extreme brightness range, the six-stop claim is this film's identity and one frame settles it.
2. A deliberate two-stop underexposure. Tests the Shadow 0 decision directly, and would vindicate the reference recipe if the shadows hold better than the sheet implies.
3. A large-detail subject at 100%. Brickwork, weave, foliage. Testing whether Sharpness +1 / Clarity 0 delivers the enlargement claim.

Derived: 2026-07-25